Disruption of Data Networks Endangers Preparedness, Anticipatory Action and Food Security

Data ecosystems are the foundation for an efficient utilization of data in a complex network of various actors, institutions and processes which jointly collect, process, store and apply such data. These humanitarian and climate data and analysis ecosystems are intertwined in more ways than many people realize. Should one of these actors or processes become unavailable, farreaching consequences can be expected with signficant disruption of decisionmaking in humanitarian, development and climate action.

What are the risks of being unable to use key data sources in the current situation of shrinking funding and with the closure of USAID?

In 2012, UN Secretary General Ban Ki-moon called for a "data revolution" and appointed an independent group of experts to promote data based develpment. In 2016, in the Grand Bargain 1.0, donors and humanitarian organizations agreed to collectively enhance anticipatory action, strengthening partnerships and enhancing needs assessments. These commitments were further affirmed as part of the Grand Bargains 2.0 (2021) and 3.0 (2023). Since the original 2016 Grand Bargain commitments, data and analysis within the humanitarian and development sectors has undergone radical change. There are now several data sources and analysis approaches, many shared on Refliefweb and the Humanitarian Data Exchange. Alongside this growth in the generation and use of data and analysis, technological advances have seen the enhancement of meteorological forecasts and the emergence of AI-assisted forecast models for displacement and conflict.

WHH and the communities we work with use these data and analysis sources daily to make a huge range of different decisions. Seasonal forecasts tell us and communities whether we need to prioritize drought tolerant seeds to mitigate a poor harvest or preposition relief items ahead of amonsoon season forecast to be wetter than normal. Food security assessments help us identify where to conduct food security and nutrition interventions. However, what often goes unseen are the invisible connections between all these sources. Our sources - from community-based to international - are intertwined. The recent demise of USAID (United States Agency for International Development) and funding cuts indicated by major donors highlight some of the fragilities across the data and analysis ecosystem we use to deliver on our core commitment to ending hunger and malnutrition worldwide.

Anticipatory Action - a data and analysis-"hungry" field

Analyze hazard, exposure, and their impacts: The Welthungerhilfe Anticipatory Humanitarian Action Facility (WAHAFA) program uses a wide range of data and analysis to facilitate timely and effective anticipatory action (AA) to mitigate the adverse impacts of forecastable hazards. To identify hazard-prone areas and understand what scale of crisis requires action, project teams use information from national disaster management agencies, international repositories of historical events (e.g. EM-DAT, part-funded by USAID) and community discussions (e.g. memories of historical events). Other event records, such as reports compiled by the UN Office for Coordination of Humanitarian Affairs (UN-OCHA) contain inputs from a range of different UN agencies, international and local NGOs, many of whom are facing funding cuts. These reports contain valuable information on historical impacts we can use to support analysis of potential hazards and where to prioritize anticipatory actions. There is no such thing as single source hazard assessment.

Assessing hazard vulnerability: With hazards and affected areas prioritized, the communities most at risk have to be identified. Hazard vulnerability is complex – we use primary data collected by our project teams from communities, blended with national and international sources, using a range of population statistics, health and socio-economic indicators. One approach for assessing vulnerability used within the WAHAFA program is ‘Household Economic Analysis (HEA)’. HEA is a framework for understanding food security, livelihoods and poverty, and is a tool for early warning. It uses a combination of primary and secondary data to delineate livelihood zones and socio-demographic divisions, access to food and livelihood incomes and determines the dependence of communities on external assistance to maintain their livelihoods and their ability to manage hazards. As with analyzing hazard exposure and their impacts, there is no single source used to determine hazard vulnerability.

Knowing when to act: We use freely available and reliable observations or forecasts to monitor emerging hazards. One example of these are seasonal climate forecasts, which use multiple forecasts from different meteorological agencies combined into a single ‘consensus forecast’. The meteorological agencies making these seasonal forecasts monitor slower-moving parts of our climate, such as fluctuating sea surface temperatures. One slow-moving example that is well monitored is the El Niño-Southern Oscillation (ENSO), which captures variations in sea surfaces temperatures in the tropical Pacific Ocean. Rising temperatures, termed the El Niño phase, alter world weather patterns, often seeing enhanced risk of floods in East Africa and drought in Southern Africa.

These warming patterns can be forecast months in advance, allowing sufficient time for anticipatory action. The United States’ National Oceanic and Atmospheric Administration (NOAA) is a leading forecaster for ENSO. ENSO forecasts made by NOAA are used by regional networks and national governments in their own Early Warning Systems and forecasts, reaching communities via various communication methods including radio stations, text messages and posters. It remains unclear how this valuable forecasting service will be affected by NOAA staffing cuts.

There is no such thing as a 100% accurate forecast. When working with uncertainty, triangulation of different forecast and observation sources, including from communities, helps us to make decisions about whether to take anticipatory actions. In January 2025, we tried to use the Famine Early Warning Systems Network (FEWS NET), a leading global provider of early warning information relating to severe food security outcomes, to triangulate reports of drought in one of our project countries. FEWS NET was offline due to the demise of USAID. Although there are some indications that the service could resume, FEWS NET remains offline at the time of writing. Without ensuring we  we risk being unable to take timely action to mitigate the impacts of forecastable hazards.

Prepositioning aid where it matters

A timely disaster response needs effective preparatory measures. One such measure is prepositioning relief stocks. To ensure crucial humanitarian supplies are available in the right place at the right time, we need data to answer two key questions:

• Who currently has what stocks in which location?
• What do historical disaster records and seasonal forecasts tell us about where those stocks might be needed in the future?

By working with data such as these from multiple sources including organizational reporting, historical hazardous event databases and hazard risk data, we can proactively preposition essential supplies before disaster strikes.

The ESUPS project (Emergency Supply Prepositioning Strategy) focuses on doing just this through the STOCK of Humanitarian Organisations Logistics Mapping (STOCKHOLM), an online platform which collects and analyses data on prepositioned relief stocks at country level. During the preparedness phase, using stock reports from humanitarian partners, combined with a mathematical model developed in partnership with Penn State University and MIT (Massachusetts Institute of Technology) , STOCKHOLM produces collaborative recommendations on defining national prepositioning strategies to inform what stock would be needed where to optimize crisis response. During the anticipation phase, STOCKHOLM will also provide specific recommendations on how to relocate relief items ahead of an event to maximise both the response efficiency and the overall stock rotation.

125 agencies from 58 countries report to the platform. This enables better coordination, encourages stock sharing, and reduces resource wastage.

Ending hunger needs data and analysis

Every year since 2006, Welthungerhilfe, Concern Worldwide and the Institution for International Law of Peace and Armed Conflict (IFHV) update the Global Hunger Index (GHI). The GHI is a tool that measures and tracks hunger at global, regional and country levels over time and is designed to trigger action to reduce hunger. The data and analysis needed to assess hunger levels reflect the multidimensional nature of hunger. The GHI utilizes four key indicators from a range of sources to reflect this complexity:

• Undernourishment: The share of the population whose caloric intake is insufficient (from FAO);
• Child stunting: The share of children under five who have low height for their age, reflecting chronic undernutrition (from WHO, UNICEF and the DHS Program);
• Child wasting: The share of children under who have low weight for their height, reflecting acute undernutrition (from WHO, UNICEF and the DHS Program);
• Child mortality: The share of children who die before their fifth birthday, reflecting in part the fatal mix of inadequate nutrition and unhealthy environments (from UN IGME).

The sources we use come from a range of UN agencies, all of whom are facing funding cuts and have seen USAID project closures. The Demographic and Health Surveys (DHS) rogram which we use to obtain data on child stunting and wasting, run by USAID, is currently on pause. Beyond the GHI, this source is essential for disseminating data on population health and nutrition worldwide and is used to inform health and nutrition programming essential for ending global hunger. As with many sources in the data and analysis ecosystem, consolidated repositories of trusted information and survey approaches provided by programs such as DHS play a valuable role in collating and analyzing data from various organizations, allowing us to systematically analyze and prioritize programming.

Another common source of information in the quest to end world hunger used by WHH across our programs is the Integrated Food Security Phase Classification (IPC). The IPC outcome, the number of people experiencing differing levels of acute food security , is informed by primary and secondary data sources from a range of organizations conducting food security, nutrition, health, WASH and livelihood assessments. The IPC also utilizes remote sensing and ground observation data to support projections of future adverse acute food security outcomes (known as projections).

The conceptual analysis framework behind the IPC reflects the causal and structural factors that lead to hunger outcomes, demonstrating the diversity of analysis required to quantify food insecurity. With the current ‘offline status’ of FEWS NET, ensuring continuity of the IPC process is essential  to identify food insecurity outcomes, and work to mitigate the next famine.

Conclusion and Future Prospects

The humanitarian, development and climate data ecosystems are highly interconnected. As the effects of the demise of USAID continue to emerge, and other major humanitarian donors look to reduce their aid budgets, it is vital that we protect our interlinked sources. In identifying at-risk communities and working to end global hunger, there is no such thing as a single, unconnected data source.

However, there are opportunities for change: In the context of hunger eradication, several opportunities exist to ensure data continuity even as funding decreases. A key opportunity lies in localizing data and analysis processes. Local actors—communities, civil society organizations, and regional authorities—are reliable sources of accurate, context-specific information. Strengthening their capacity to collect and analyze data ensures that critical information continues to be gathered during challenging times. Additionally, integrating local networks into tools like the GHI enhances the quality and relevance of the data. This approach not only improves data security but also enables more adaptive and sustainable hunger eradication efforts that are closely aligned with the needs and realities on the ground.

In Anticipatory Action, localization plays an equally crucial role. Incorporating local observations and knowledge into analysis increases the accuracy of predictions on where future hazards are likely to occur. Also, strengthened advocacy for ‘no regrets’ approaches when working with uncertainty ensures that we will be able to facilitate timely action in the face of potential risks to forecasting sources. No regrets approaches recognize that anticipatory actions are beneficial for resilience and development gains, regardless of whether the hazard event emerges as forecast.

Local actors can contribute to data collection and play an essential role in prepositioning relief supplies and developing national strategies. Ultimately, by putting communities and local actors at the heart of the data and analysis ecosystem we can ensure our data and analysis systems are grounded in reality, inclusive and accessible, supporting our goal to end global hunger.